Cost of Care Coordination for Children With Special Health Care Needs

Initial experiences in managed care have demonstrated that disease management approaches that incorporate targeted case management and quality management in addition to physician training and outreach can significantly improve quality and reduce the costs of care for certain populations with chronic illness. These efforts have initially been focused on high-incidence or high-cost groups such as those persons with diabetes, asthma, depression, or high-risk pregnancy. Some programs address children with chronic illness. For example, Greineder et al¹ demonstrated that selected high-risk children with asthma who were enrolled in an outreach program operated by the Harvard Community Health Plan, Boston, Mass, achieved reduced rates of hospitalization, reduced rates of emergency department utilization, and improved quality of care. The program used a registered nurse working approximately 8 hours per week, which represented an annual cost of $11,115 in 1993. The program generated a savings of $76,200 per year or an almost 8:1 ratio of savings to costs.

We are interested in the utility of case management and disease management for more heterogeneous populations enrolled in mandatory state Medicaid managed care programs. Our research has demonstrated that a targeted case management and disease management program for children and adults with developmental disabilities, operating in a fee-for-service environment, can achieve substantial reductions in inpatient services utilization. We found that inpatient utilization for this population can be reduced by approximately 25%^{2 - 4} through similar interventions. However, we have not been able to demonstrate a net cost savings as the costs of the program are roughly equal to savings resulting from decreased use of inpatient resources.

With this in mind, we were quite interested in the recent report by Liptak et al,⁵ which demonstrated that a population-based, targeted case management and disease management approach for a heterogeneous population such as children with chronic illness could achieve cost savings comparable to those seen in more focused interventions. The authors describe a collaboration between the Children's Hospital at Strong in Rochester, NY, and a regional insurance company. The insurance company provided funds to the hospital that allowed for the hiring of 11 additional full-time equivalents—including nurses, social workers, psychologists, nutritionists, physical therapists, occupational therapists, special educators, play therapists, and speech and language pathologists—to expand ambulatory care coordination and "wrap-around" services to children with chronic conditions. There were several important findings in the study. First, the authors found that physicians and other health-related workers spent 44% of their clinical time in encounters that were not billable under most fee-for-service plans. This finding points to the need for creative financing arrangements that can support case management and other wrap-around services. Second, the study demonstrated cost savings of $10.50 for every $1 invested in the program. The authors correctly identify several limitations with the generalizability of their findings, including the potential for cost shifting to outpatient programs, rehabilitation and educational programs, and families. However, we were uncertain if confounding variables, including the effects of statistical outliers or the role of managed care, may have contributed to the health care savings attributed by the authors to the case management program.

First, our research on length of stay for persons with developmental disabilities demonstrated that a few patients can have a profound effect on the data. In our analysis of 428 hospital admissions during a 3-year period, we found significant differences between the length of stay for intervention (3.81 days) and control groups (8.60 days) (P=.38). However, these findings were demonstrated after removing 4 admissions that accounted for a total of 486 inpatient days. We felt that this was prudent given the relatively small size of the sample and the fact that these outliers often had nonmedical or disposition issues associated with their lengthy stays. Liptak et al⁵ state that the mean length of stay for children with chronic conditions admitted to the Children's Hospital at Strong decreased from 83.9 to 10.6 days. On face value, these lengths of stay seem inordinately long. We are curious as to whether a portion of this length of stay was related to a small number of individuals and whether the authors removed outliers from their data set before conducting their statistical analyses.

Second, the article also suggests that the market dynamics within the region have been changing. For example, their Figure 1 demonstrates that nearly all of the change in the number of admissions reported between 1983 and 1995 occurred between 1989 and 1991. This was accompanied by an increase in managed care penetration. The authors note that "in 1989, a capitation agreement was made with a regional insurance company in the Rochester area to support the annual costs of this program."⁵ (p1004) We wondered if the cost savings were due to the greater penetration of managed care in the local market. If so, this could have affected the financial incentives of the health care providers in the region. Did the hospital enter into a capitation agreement with the insurer for the coverage of inpatient care? Similarly, did the insurance company enter into capitation agreements with the physician groups who admitted patients to the hospital during the study? Finally, what degree of integration exists between the medical school, its faculty, and the community physicians? Were new programs such as specialized nursing facilities or other long-term care options created? Answers to these questions could be useful to centers that are considering replication of this model and could have significantly affected the conclusion of the authors that the cost savings should be attributed solely to the case management program.

We applaud the researchers for their effort in undertaking this line of research. Like most clinicians who care for children with chronic illness, we are struggling with the burden of providing care for which we are uncompensated. The experience at Children's Hospital at Strong helps place our own experience in perspective and suggests new models of contracting. In this light, we hope our gentle inquiries will increase the likelihood that similar programs can be created in other settings.